Method of classifying membership categories

Abstract

The present invention discloses a method of classifying membership categories. The method comprises: determining a time starting point, creating a total database on the basis of member behavior information before the time starting point, creating an incremental database on the basis of member behavior information at a period of time after the time starting point, executing a process of comparing member data of the total database and the incremental database, judging whether there is the same member code data in the total database as that in the incremental database, updating the total database on the basis of the judging result, obtaining a final total database, and classifying membership categories. The present invention avoids the problem of huge expenses required for reading fact table partition data each time in batches by creating a total database and updating the total database in real-time through the mode of generating incremental database in real-time.

Description

METHOD OF CLASSIFYING MEMBERSHIP CATEGORIES BACKGROUND OF THE INVENTION Technical Field

[0001] The present invention pertains to the field of big data oflline processing, and particularly relates to a method of classifying membership categories. Description of Related Art

[0002] Currently available classification of membership categories is mainly based on behavior data generated by members, namely to extract member numbers, purchasing and browsing behavior date data, purchasing and browsing behavior fact frequencies in the behavior information data, and then to classify membership categories on the basis of such data. However, such method is operationally defective in two aspects: on one hand, member behavior information contains various types of data, while not every type of data is suitable for classification of membership categories, if all data of the member behavior information is directly extracted for the classification of membership categories, a great deal of working load would be put on the processor; on the other hand, if membership categories are classified on the basis of common member behavior information data, the behavior information data of all members should be repetitively read each time, and this would engender a great deal of repetitive work contents. SUMMARY OF THE INVENTION

[0003] An objective of the present invention rests in solving the aforementioned problem existent in classifying membership categories, to hence propose a method of creating a total database and a method of realizing classification of membership categories on the basis of the total database. Calculations performed to labels such as old category, new category, 1 Date re9ue I Date received 2021-11-08 retaining category and repeatedly purchasing category, etc. of members on the basis of this method enable to avoid the problem of huge expenses required for reading fact table partition data each time in great batches, to reduce the overhead for aggregated calculation of the data in great batches to a larger extent, and to economize on the computational cost to a greater degree.

[0004] The specific method comprises:

[0005] determining a time starting point, selecting all member behavior information before the time starting point, and creating a total database;

[0006] determining a time cutoff point, selecting all member behavior information between the time starting point and the time cutoff point, and creating an incremental database, wherein the time starting point is prior to the time cutoff point, and both the total database and the incremental database at least include member code data;

[0007] respectively traversing in the total database all member code data in the incremental database, updating the total database according to the traversing result, obtaining a final total database; and classifying membership categories on the basis of the final total database.

[0008] Preferably, the step of respectively traversing in the total database all member code data in the incremental database includes:

[0009] executing a comparison process of the member code data, selecting one piece of member code data from the incremental database, traversing the total database, and judging whether this piece of member code data exists in the total database;

[0010] if this piece of member code data exists in the total database, merging member data to 2 Date re9ue I Date received 2021-11-08 which this piece of member code data corresponds in the incremental database with member data to which this piece of member code data corresponds in the total database, and updating the total database;

[0011] if this piece of member code data does not exist in the total database, adding to the total database this piece of member code data and member data to which this piece of member code data corresponds, and updating the total database; and

[0012] sequentially and repeatedly executing the comparison process of remaining member code data in the incremental database, until all member code data in the incremental database have been compared with that of the total database.

[0013] Preferably, the method further comprises updating the time starting point and the time cutoff point; taking the time cutoff point as a new time starting point, and selecting a new time cutoff point after the time cutoff point; selecting all member behavior information between the new time starting point and the new time cutoff point, creating an incremental database; and executing the comparison process of member data.

[0014] Preferably, the member data to which the member code data corresponds further at least includes behavior date data and behavior fact value of the member, and the behavior date data and the behavior fact value correspond to each other.

[0015] Preferably, the step of merging member data to which this piece of member code data corresponds in the incremental database with member data to which this piece of member code data corresponds in the total database, and updating the total database further includes:

[0016] traversing a piece of behavior date data to which a piece of member code data corresponds in the incremental database in member data to which this piece of member code data corresponds in the total database; 3 Date re9ue I Date received 2021-11-08

[0017] if this piece of behavior date data exists in the member data to which the member code data corresponds in the total database, replacing a behavior fact value to which the behavior date data corresponds, to which the member code data corresponds in the total database with a behavior fact value to which the behavior date data corresponds, to which the member code data corresponds in the incremental database, and updating the total database;

[0018] if this piece of behavior date data does not exist in the member data to which the member code data corresponds in the total database, adding the behavior date data to which the member code data corresponds and the behavior fact value to which the behavior date data corresponds in the incremental database to the member data to which the member code data corresponds in the total database, and updating the total database; and

[0019] sequentially traversing other behavior date data to which the member code data corresponds in the incremental database in the member data to which the member code data corresponds in the total database, until all behavior date data to which the member code data corresponds in the incremental database have been traversed.

[0020] There is provided a device for classifying membership categories, and the device preferably comprises:

[0021] a database creating module, for determining a time starting point, selecting all member behavior information before the time starting point, creating a total database, determining a time cutoff point, selecting all member behavior information between the time starting point and the time cutoff point, and creating an incremental database, wherein the time starting point is prior to the time cutoff point, and both the total database and the incremental database at least include member code data; and

[0022] a data comparing module, for respectively traversing in the total database all member 4 Date re9ue I Date received 2021-11-08 code data in the incremental database, updating the total database according to the traversing result, obtaining a final total database, and classifying membership categories on the basis of the final total database.

[0023] Preferably, the data comparing module further includes:

[0024] a first judging unit, for executing a comparison process of the member code data, selecting one piece of member code data from the incremental database, traversing the total database, and judging whether this piece of member code data exists in the total database; and

[0025] a first data updating unit, for updating the total database on the basis of a judging result of the first judging unit.

[0026] Preferably, the first data updating unit further includes:

[0027] a second judging unit, for judging whether a piece of behavior date data to which the member code data corresponds in the incremental database exists in the member data to which the member code data corresponds in the total database; and

[0028] a second data updating unit, for updating the total database on the basis of a judging result of the second judging unit, and sequentially traversing other behavior date data to which the member code data corresponds in the incremental database in the member data to which the member code data corresponds in the total database, until all behavior date data to which the member code data corresponds in the incremental database have been traversed.

[0029] There is provided a computer equipment that comprises a memory, a processor and a computer program stored in the memory and executable on the processor. Preferably, the 5 Date re9ue I Date received 2021-11-08 aforementioned method is realized when the processor executes the computer program.

[0030] There is provided a computer-readable storage medium storing thereon a computer program. Preferably, any of the aforementioned methods is realized when the computer program is executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] To more clearly describe the technical solutions in the embodiments of the present invention, drawings used to illustrate the embodiments are briefly introduced below. Apparently, the drawings introduced below are merely directed to some embodiments of the present invention, and persons ordinarily skilled in the art may acquire other drawings from these drawings without spending creative effort in the process.

[0032] Fig. 1 is a :flowchart illustrating singly updating a total database and classifying membership categories according to the present invention; and

[0033] Fig. 2 is a :flowchart illustrating consistently updating the total database and classifying membership categories according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0034] To make the objectives, technical solutions and advantages of the present invention more lucid and clearer, the technical solutions in the embodiments of the present invention will be clearly and comprehensively described below with reference to the embodiments of the present invention. Apparently, the described embodiments are merely partial, rather than the entire, embodiments of the present invention. All other embodiments achievable by persons ordinarily skilled in the art on the basis of the embodiments of the present invention without spending creative effort shall all fall within the protection scope of the present invention. 6 Date re9ue I Date received 2021-11-08

[0035] As should be noted, unless otherwise explicitly required in the context, such wordings as "including" and "comprising" used throughout this Description and the attached Claims should be explained as denoting inclusion rather than exclusion or exhaustion, in other words, they denote the meaning of"including, but not being limited to ... ".

[0036] As should be additionally noted, unless otherwise explained, such wording as "a plurality of' and "plural" used in the description of the present invention denotes the meaning of "two or more".

[0037] As noted in the Description of Related Art, classification of membership categories is executed mainly on the basis of great amount of member behavior information, this is an extremely onerous data processing event, and the operation of which should be repeatedly executed with the passing of time.

[0038] Embodiment One

[0039] An embodiment of the present invention proposes a method of realizing classification of membership categories, and the method can comprise the following steps: S 1. creating a total database and an incremental database.

[0040] S 11. creating a total database. A time starting point is determined, and the total database is created on the basis of all member behavior information before the time starting point. The total database at least includes member data, and the member data is sequenced according to member code data. The member data at least includes member code data, behavior date data and behavior fact values.

[0041] In order to make more convenient for the total database to subsequently classify 7 Date re9ue I Date received 2021-11-08 membership categories, a corresponding logical relation is established among the member code data, behavior date data and behavior fact values in the total database through JSON array. That is, [ member code data, {behavior date data, behavior fact values}], in which a behavior fact value indicates the frequency of such behaviors as the user browses a commodity, or puts an order on a commodity, or lists a commodity on a wish list; described below is an example in which the user put an order on a commodity, a total database was created at the time node 2020, October 20, and all order data before 2020, October 20 is extracted; for instance, a user with member code data as 00001 put an order on two commodities on 2020, October 1, and put an order on one commodity on 2020, October 3, then these would be displayed in the total database as [00001, {20201001, 2}, {20201003, 1} ]l, all data are sequenced from top to bottom according to the member code data, and the database as formed is shown below: [00001, {20201001,2}, {20201003, l}] [00002, {20110701, 2}] [00003, {20130811,2}, {20140902,2}, {20201003, l}] [00004, {20140902,2}, {20201003,4}]

[0042] S12. Creating an incremental database and determining a time cutoff point, selecting all member behavior information between the time starting point and the time cutoff point to create the incremental database, wherein the incremental data base also at least includes member data, and the member data at least includes member code data, behavior date data and behavior fact values.

[0043] The formats of the member code data, behavior date data and behavior fact values in the incremental database are the same as those formats in the total database; take for example the data between 2020, October 20 and 2020, October 25, in which a user with the member code data as 00001 put an order on a commodity on 2020, October 22, and put an order on 8 Date re9ue I Date received 2021-11-08 two commodities on 2020, October 23, then these would be displayed in the incremental database as [00001, {20201022, 1}, {20201023, 2} ], and a user with the member code data as 00088 put an order on a commodity on 2020, October 22, and put an order on a commodity on 2020, October 23, then these would be displayed in the incremental database as [00088, {20201022, l}, {20201023, l}]. All data are sequenced from top to bottom according to the member code data, and the database as formed is shown below: [00001, {20201022, l}, {20201023,2}] [00088, {20201022, l}, {20201023, l}]

[0044] S2. updating the total database on the basis of the incremental database.

[0045] The incremental database is compared with the total database to judge whether the member code data in the incremental database exists in the total database: if not, the member code data and such information as the behavior date data and the behavior fact values to which the member code data corresponds in the incremental database are updated into the total database; if yes, such information as the behavior date data and the behavior fact values to which the member code data corresponds in the incremental database is updated into the same location where the member code data corresponds in the total database.

[0046] The above data is taken for example to enunciate the comparison process of the incremental database with the total database; the information data between 2020, October 20 and 2020, October 25 of the incremental database is updated into the total database created before 2020, October 20, and each member code data in the incremental database is sequentially traversed in the total database, that is, the 00001 member code data in the incremental database is firstly traversed in the total database, and it is found after traversing 9 Date re9ue I Date received 2021-11-08 that the 00001 member code data also exists in the total database, then such information {20201022, l}, {20201023, 2}] as the behavior date data and the behavior fact values to which the member code data corresponds in the incremental database is updated into the same location where the member code data corresponds in the total database, i.e., the fields of the 00001 member code data in the updated total database are: [00001, {20201001, 2}, {20201003, l}, {20201022, l}, {20201023, 2} ], and the remaining member code data are sequentially traversed, for instance, the 00888 member code data is traversed, while there is no 00888 member code data in the total database, then the relevant fields of the 00888 member code data are added to the total database; by traversing all member code data in the incremental database, a new total database is finally obtained, as follows: [00001, {20201001,2}, {20201003, l}, {20201022, l}, {20201023,2}] [00002, {20110701,2}] [00003, {20130811,2}, {20140902,2}, {20201003, l}] [00004, {20140902,2}, {20201003,4}] [00088, {20201022, l}, {20201023, l}]

[0047] S3. Continuously updating the new total database in real time with the passing of time. The updating process and steps are as the following example. [ 0048] S31. Updating time starting point and time cutoff point, changing the time cutoff point into a new time starting point, and selecting a new time cutoff point after the new time starting point. In this embodiment for example, the new time starting point is 2020, October 25, and, to more conveniently explain this embodiment, the new time cutoff point is determined as 2020, October 30.

[0049] S32. Creating a new incremental database, as follows: 10 Date re9ue I Date received 2021-11-08 [00002, {20201027,2}, {20201028,2}] [00003, {20201003,3}] [00089, {20201027, l}]

[0050] It is found during the process of creating this new incremental database that the number of orders of member code 00003 is not completely counted at 2020, October 3 due to data delay in the previous process of creating the total database, and the count is complete only until the time period from 2020, October 25 to 2020, October 30. That is to say, three orders were generated by member code 00003 by 2020, October 3, but only one order is counted.

[0051] Based on the above finding, the data of the member code at 2020, October 3 is still counted in the new incremental database, and the total database is updated.

[0052] S33. Continuously updating the total database according to the new incremental database. As should be noted, during the updating process, behavior fact value 1 in [00003, {20201003, 1}] in the total database is directly replaced with behavior fact value 3 in [00003, {20201003, 3}] in the new incremental database to obtain a total database, as follows: [00001, {20201001,2}, {20201003, l}, {20201022, l}, {20201023,2}] [00002, {20110701,2}, {20201027,2}, {20201028,2}] [00003, {20130811,2}, {20140902,2}, {20201003,3}] [00004, {20140902,2}, {20201003,4}] [00088, {20201022, l}, {20201023, l}] [00080, {20201027, l}] Date re9ue I Date received 2021-11-08 11

[0053] S4. Assigning a label to the user according to the final total database. For instance, a user with the member code data as 00001, 00002, 00003, 00004, 00088 ...... exhibits plural ordering behaviors, and she or he is defined as a repeatedly purchasing member; a user with the member code data as 00001, 00002, 00003, 00004 ...... exhibits more than one ordering behavior everyday, and she or he is defined as a high consumption member; a user with the member code data as 00089 ...... exhibits only one ordering behavior, and she or he is defined as a silent member.

[0054] Embodiment Two

[0055] An embodiment of the present invention provides a device for classifying membership categories, and the device comprises a database creating module that is capable of selecting all order data before a certain time point, extracting valid data from the order data, and creating a total database.

[0056] For instance, the order data includes member code data, behavior date data, behavior fact values, commodity code data, shop code data, etc.

[0057] The database creating module can select orders prior to 2020, October 1, remove such invalid data as commodity code data and shop code data, and only extract member code data, behavior date data, and behavior fact values.

[0058] After the total database and the incremental database have been created by the database creating module, a data comparing module can compare the information in the incremental database with the information in the total database, to thereby update the total database.

[0059] In order to realize the process of comparing data, the data comparing module includes a first judging unit and a first data updating unit, the task executed by the first judging unit 12 Date re9ue I Date received 2021-11-08 is to judge whether each member code data in the incremental database exists in the total database, and the first data updating unit bases on the existing result to update the first total database. The first data updating unit further includes a second judging unit and a second data updating unit. The second judging unit is employed for judging whether all behavior date data to which the member code data corresponds in the incremental data exists in the member data to which the member code data corresponds in the total database, and the second data updating unit is employed for updating the total database on the basis of the judging result of the second judging unit.

[0060] For instance, an existing total database is as follows: [00001, {20201001,2}, {20201003, l}] [00002, {20110701,2}] [00003, {20130811,2}, {20140902,2}, {20201003, l}] [00004, {20140902,2}, {20201003,4}]

[0061] The incremental database is as follows: [00001, {20201003,2}, {20201022, l}, {20201023,2}]

[00088] , {20201022, l}, {20201023, l}

[0062] The first judging unit traverses the member code data 00001 and 00088 in the total database to determine whether the member code data 00001 and 00088 exists in the member codes of the total database. In this example, the member code data 00001 in the incremental database exists in the total database, while the member code data 00088 does not exist in the total database. The first data updating unit then, on the basis of the judging result of the first judging unit, updates the member behavior date data and member behavior fact values to which the member code data 00088 corresponds into the total database, to preliminarily obtain a total database: 13 Date re9ue I Date received 2021-11-08 [00001, {20201001,2}, {20201003, l}] [00002, {20110701,2}] [00003, {20130811,2}, {20140902,2}, {20201003, l}] [00004, {20140902,2}, {20201003,4}] [00088, {20201022, l}, {20201023, l}]

[0063] The second judging unit judges whether all behavior date data 20201003, 20201022, 20201023 to which the member code data 00001 corresponds in the incremental database exists in the fields to which the member code data 00001 corresponds in the total database, and the judging result is that the behavior date data 20201022 and 20201023 do not exist in the fields to which the member code data 00001 corresponds in the total database, and that the behavior date data 20201003 exists in the fields to which the member code data 00001 corresponds in the total database. The second data updating unit executes the process of updating the total database by adding the behavior date data 20201022 and 20201023 to which the member code data 00001 corresponds and the behavior fact values to which the behavior date data corresponds to the fields to which the member code data 00001 corresponds in the total database, and replacing the behavior fact value to which the behavior date data 20201003 corresponds to which the member code data 00001 in turn corresponds in the incremental database with behavior fact value 2 to which the behavior date data 20201003 corresponds to which the member code data 00001 in turn corresponds in the total database. A final total database is obtained by updating, as follows: [00001, {20201001,2}, {20201003,2}, {20201022, l}, {20201023,2}] [00002, {20110701,2}] [00003, {20130811,2}, {20140902,2}, {20201003, l}] [00004, {20140902,2}, {20201003,4}] [00088, {20201022, l}, {20201023, l}] 14 Date re9ue I Date received 2021-11-08

[0064] The aforementioned method can be realized when a processor executes a computer program.

[0065] There is provided a computer-readable storage medium storing thereon a computer program, and the aforementioned method is realized when the computer program is executed by a processor.

[0066] To sum it up, the advantageous effect achievable by the technical solutions provided by the embodiments of the present invention is that, by means of the scenario identifying method for the classification of membership categories, the corresponding device, computer program and storage medium, a method of creating a total database and a method of realizing classification of membership categories on the basis of the total database are provided. Calculations oflabels such as old category, new category, retaining category and repeatedly purchasing category, etc. of members on the basis of these methods enable to avoid the problem of huge expenses required for reading fact table partition data each time in great batches, to reduce the overhead for aggregated calculation of the data in great batches to a larger extent, and to economize on the computational cost to a greater degree.

[0067] As should be noted, it can be understood by persons ordinarily skilled in the art that the entire or partial steps of the aforementioned embodiments can be completed via hardware, or by a program that instructs relevant hardware, that the program can be stored in a computer-readable storage medium, and that the storage medium can be a read-only memory, a magnetic disk or an optical disk, etc.

[0068] What is described above is merely directed to preferred embodiments of the present invention, and is not meant to restrict the present invention. Any amendment, equivalent replacement and improvement makeable within the spirit and principle of the present invention shall all be covered within the protection scope of the present invention. 15 Date re9ue I Date received 2021-11-08

Claims

16 Claims:

1. A device for classifying membership categories, the device comprising: a database creating module configured to: determine a starting point; select member behavior information before the starting point; create a first database on the basis of the member behavior information before the starting point; determine a cutoff point; select member behavior information between the starting point and the cutoff point; create a second database on the basis of the member behavior information between the starting point and the cutoff point; wherein the starting point is prior to the cutoff point, and wherein the first database and the second database comprise member code data and member data sequenced according to the member code data; and a data comparing module configured to: respectively traverse each piece of the member code data from the second database in the first database to generate a traversing result; update the first database according to the traversing result; obtain a third database; and classify membership categories on the basis of the third database. CA 3138147 Date reçue / Received date 2025-11-09 17 2. The device of claim 1, wherein the data comparing module classifies the membership categories on the basis of the third database by assigning a label to a user according to the third database.

3. The device of claim 2, wherein the label is “old category”.

4. The device of claim 2, wherein the label is “new category”.

5. The device of claim 2, wherein the label is “retaining category”.

6. The device of claim 2, wherein the label is “repeatedly purchasing category”.

7. The device of any one of claims 1-6, wherein the data comparing module further comprises: a first judging unit configured to: execute a comparison of the member code data; select each piece of member code data from the second database; traverse the first database; judge whether the piece of member code data exists in the first database; and generate a first judging result; and a first data updating unit for updating the first database on the basis of the first judging result.

8. The device of claim 7, wherein the first data updating unit is configured to: where the piece of member code data exists in the first database, merge member data to which the piece of member code data corresponds in the second database with member data to which the piece of member code data corresponds in the first database; where the piece of member code data does not exist in the first database, add to the first database the piece of member code data and the member data to which the piece of member code data corresponds and update the first database. CA 3138147 Date reçue / Received date 2025-11-09 18 9. The device of claim 8, wherein where the piece of member code data exists in the first database, the first data updating unit is further configured to add the member data to attribute fields of the first database.

10. The device of claim 8, wherein where the piece of member code data does not exist in the first database, the first data updating unit is further configured to: add the member data to primary key and attribute fields of the first database; and record and store the member code.

11. The device according to claim 7, wherein the device is further configured to: update the starting point and the cutoff point; take the cutoff point as a new starting point, and selecting a new cutoff point after the cutoff point; select member behavior information between the new starting point and the new cutoff point; create a new second database on the basis of the member behavior information between the new starting point and the new cutoff point; and execute the comparison of the member data.

12. The device of any one of claims 1 to 11, wherein the device is further configured to judge whether member codes in the second database have been stored.

13. The device of any one of claims 1 to 12, wherein the member data to which the member code data corresponds comprises behavior date data and behavior fact value of the member, and wherein the behavior date data and the behavior fact value mutually correspond.

14. The device of claim 13, wherein the member data to which the member code data corresponds comprises commodity code data and / or shop code data. CA 3138147 Date reçue / Received date 2025-11-09 19 15. The device of any one of claims 13 to 14, wherein the device is further configured to extract valid data from the member behavior information.

16. The device of claim 15, wherein the extracted valid data comprises the member code data, the behavior date data, and the behavior fact value.

17. The device of any one of claims 13-16, wherein a corresponding logical relation is established among the member code data, the behavior date data and the behavior fact value in the first database through a JSON array.

18. The device of any one of claims 13 to 17, wherein each behavior fact value indicates the frequency of a corresponding behavior as a user browses a commodity, puts an order on the commodity, or lists the commodity on a wish list.

19. The device of any one of claims 1 to 18, wherein the first data updating unit further comprises: a second judging unit configured to: judge whether each piece of behavior date data to which the member code data corresponds in the second database exists in the member data to which the member code data corresponds in the first database; and generate a second judging result; and a second data updating unit configured to: update the first database on the basis of the second judging result.

20. The device according to claim 19, wherein the second data updating unit is further configured to: where the piece of behavior date data exists in the member data to which the member code data corresponds in the first database, replace a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the first database, CA 3138147 Date reçue / Received date 2025-11-09 20 with a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the second database; where the piece of behavior date data does not exist in the member data to which the member code data corresponds in the first database, add the behavior date data to which the member code data corresponds and the behavior fact value to which the behavior date data corresponds in the second database to the member data to which the member code data corresponds in the first database.

21. A device for classifying membership categories, the device comprising: a processor; a memory, associated with the processor, configured to store a program instruction that, when executed by the processor, causes the processor to perform the following operations: determining a starting point; selecting member behavior information from before the starting point; creating a first database on the basis of the member behavior information from before the starting point; determining a cutoff point; selecting member behavior information from between the starting point and the cutoff point; creating a second database on the basis of the member behavior information between the starting point and the cutoff point; wherein the starting point is prior to the cutoff point, and wherein the first database and the second database each comprise member code data and member data sequenced according to the member code data; CA 3138147 Date reçue / Received date 2025-11-09 21 respectively traversing each piece of the member code data from the second database in the first database to generate a traversing result; updating the first database according to the traversing result; obtaining a third database; and classifying membership categories on the basis of the third database.

22. The device of claim 21, wherein classifying the membership categories on the basis of the third database comprises assigning a label to a user according to the third database.

23. The device of claim 22, wherein the label is “old category”.

24. The device of claim 22, wherein the label is “new category”.

25. The device of claim 22, wherein the label is “retaining category”.

26. The device of claim 22, wherein the label is “repeatedly purchasing category”.

27. The device according to any one of claims 21-26, wherein respectively traversing member code data from the second database in the first database comprises: executing a comparison of the member code data; selecting each piece of member code data from the second database; traversing the first database; and judging whether the piece of member code data exists in the first database.

28. The device according to claim 23, wherein updating the first database comprises: where the piece of member code data exists in the first database, merging member data to which the piece of member code data corresponds in the second database with member data to which the piece of member code data corresponds in the first database; CA 3138147 Date reçue / Received date 2025-11-09 22 where the piece of member code data does not exist in the first database, adding to the first database the piece of member code data and the member data to which the piece of member code data corresponds.

29. The device according to claim 28, wherein merging the member data to which the piece of member code data corresponds in the second database with member data to which the piece of member code data corresponds comprises adding the member data to attribute fields of the first database.

30. The device according to claim 28, wherein adding to the first database the piece of member code data and the member data to which the piece of member code data corresponds comprises: adding the member data to primary key and attribute fields of the first database; and recording and storing the member code.

31. The device according to claim 22, wherein the operations further comprise: updating the starting point and the cutoff point; taking the cutoff point as a new starting point, and selecting a new cutoff point after the cutoff point; selecting member behavior information between the new starting point and the new cutoff point; creating a new second database on the basis of the member behavior information between the new starting point and the new cutoff point; and executing the comparison of the member data.

32. The device according to any one of claims 21 to 30 further comprising judging whether member codes in the second database have been stored. CA 3138147 Date reçue / Received date 2025-11-09 23 33. The device according to claim 22, wherein the member data to which the member code data corresponds comprises behavior date data and behavior fact value of the member, and wherein the behavior date data and the behavior fact value mutually correspond.

34. The device according to claim 22 or 24, wherein the member data to which the member code data corresponds comprises commodity code data and / or shop code data.

35. The device of claim 33 further comprising extracting valid data from the member behavior information.

36. The device of claim 35, wherein the extracted valid data comprises the member code data, the behavior date data, and the behavior fact value.

37. The device according to claim 36, wherein a corresponding logical relation is established among the member code data, the behavior date data and the behavior fact value in the first database through a JSON array.

38. The device according to claim 37, wherein each behavior fact value indicates the frequency of a corresponding behavior as a user browses a commodity, puts an order on the commodity, or lists the commodity on a wish list.

39. The device according to claim 28, wherein merging the member data to which the piece of member code data corresponds in the second database with the member data to which the piece of member code data corresponds in the first database and updating the total database comprises: traversing each piece of behavior date data to which the piece of member code data corresponds in the second database in member data to which the piece of member code data corresponds in the first database.

40. The device according to claim 29, wherein updating the first database comprises: where the piece of behavior date data exists in the member data to which the member code data corresponds in the first database, replacing a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the first database, CA 3138147 Date reçue / Received date 2025-11-09 24 with a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the second database and updating the first database; where the piece of behavior date data does not exist in the member data to which the member code data corresponds in the first database, adding the behavior date data to which the member code data corresponds and the behavior fact value to which the behavior date data corresponds in the second database to the member data to which the member code data corresponds in the first database and updating the first database.

41. A system for classifying membership categories, the system comprising: a database creating server configured to: determine a starting point; select member behavior information before the starting point; create a first database on the basis of the member behavior information before the starting point; determine a cutoff point; select member behavior information between the starting point and the cutoff point; create a second database on the basis of the member behavior information between the starting point and the cutoff point; wherein the starting point is prior to the cutoff point, and wherein the first database and the second database comprise member code data and member data sequenced according to the member code data; and a data comparing server configured to: respectively traverse each piece of the member code data from the second database in the first database to generate a traversing result; CA 3138147 Date reçue / Received date 2025-11-09 25 update the first database according to the traversing result; obtain a third database; and classify membership categories on the basis of the third database.

42. The system of claim 41, wherein the data comparing server classifies the membership categories on the basis of the third database by assigning a label to a user according to the third database.

43. The system of claim 42, wherein the label is “old category”.

44. The system of claim 42, wherein the label is “new category”.

45. The system of claim 42, wherein the label is “retaining category”.

46. The system of claim 42, wherein the label is “repeatedly purchasing category”.

47. The system of any one of claims 41-46, wherein the data comparing server further comprises: a first judging unit configured to: execute a comparison of the member code data; select each piece of member code data from the second database; traverse the first database; judge whether the piece of member code data exists in the first database; and generate a first judging result; and a first data updating unit for updating the first database on the basis of the first judging result.

48. The system of claim 47, wherein the first data updating unit is configured to: CA 3138147 Date reçue / Received date 2025-11-09 26 where the piece of member code data exists in the first database, merge member data to which the piece of member code data corresponds in the second database with member data to which the piece of member code data corresponds in the first database; where the piece of member code data does not exist in the first database, add to the first database the piece of member code data and the member data to which the piece of member code data corresponds and update the first database.

49. The system of claim 48, wherein where the piece of member code data exists in the first database, the first data updating unit is further configured to add the member data to attribute fields of the first database.

50. The system of claim 48, wherein where the piece of member code data does not exist in the first database, the first data updating unit is further configured to: add the member data to primary key and attribute fields of the first database; and record and store the member code.

51. The system according to claim 47, wherein the system is further configured to: update the starting point and the cutoff point; take the cutoff point as a new starting point, and select a new cutoff point after the cutoff point; select member behavior information between the new starting point and the new cutoff point; create a new second database on the basis of the member behavior information between the new starting point and the new cutoff point; and execute the comparison of the member data.

52. The system of any one of claims 41 to 51, wherein the system is further configured to judge whether member codes in the second database have been stored. CA 3138147 Date reçue / Received date 2025-11-09 27 53. The system of any one of claims 41 to 52, wherein the member data to which the member code data corresponds comprises behavior date data and a behavior fact value of the member, and wherein the behavior date data and the behavior fact value mutually correspond.

54. The system of claim 53, wherein the member data to which the member code data corresponds comprises commodity code data and / or shop code data.

55. The system of any one of claims 53 to 54, wherein the device is further configured to extract valid data from the member behavior information.

56. The system of claim 55, wherein the extracted valid data comprises the member code data, the behavior date data, and the behavior fact value.

57. The system of any one of claims 53 to 56, wherein a corresponding logical relation is established among the member code data, the behavior date data and the behavior fact value in the first database through a JSON array.

58. The system of any one of claims 53 to 57, wherein each behavior fact value indicates the frequency of a corresponding behavior as a user browses a commodity, puts an order on the commodity, or lists the commodity on a wish list.

59. The system of any one of claims 41 to 58, wherein the first data updating unit further comprises: a second judging unit configured to: judge whether each piece of behavior date data to which the member code data corresponds in the second database exists in the member data to which the member code data corresponds in the first database; and generate a second judging result; and a second data updating unit configured to: update the first database on the basis of the second judging result. CA 3138147 Date reçue / Received date 2025-11-09 28 60. The system according to claim 59, wherein the second data updating unit is further configured to: where the piece of behavior date data exists in the member data to which the member code data corresponds in the first database, replace a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the first database, with a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the second database; where the piece of behavior date data does not exist in the member data to which the member code data corresponds in the first database, add the behavior date data to which the member code data corresponds and the behavior fact value to which the behavior date data corresponds in the second database to the member data to which the member code data corresponds in the first database.

61. A method of classifying membership categories, the method comprising: determining a starting point; selecting member behavior information from before the starting point; creating a first database on the basis of the member behavior information from before the starting point; determining a cutoff point; selecting member behavior information from between the starting point and the cutoff point; creating a second database on the basis of the member behavior information from between the starting point and the cutoff point; CA 3138147 Date reçue / Received date 2025-11-09 29 wherein the starting point is prior to the cutoff point, and wherein the first database and the second database each comprise member code data and member data sequenced according to the member code data; respectively traversing each piece of the member code data from the second database in the first database to generate a traversing result; updating the first database according to the traversing result; obtaining a third database; and classifying membership categories on the basis of the third database.

62. The method of claim 61, wherein classifying the membership categories on the basis of the third database comprises assigning a label to a user according to the third database.

63. The method of claim 62, wherein the label is “old category”.

64. The method of claim 62, wherein the label is “new category”.

65. The method of claim 62, wherein the label is “retaining category”.

66. The method of claim 62, wherein the label is “repeatedly purchasing category”.

67. The method according to any one of claims 61-66, wherein respectively traversing member code data from the second database in the first database comprises: executing a comparison of the member code data; selecting each piece of member code data from the second database; traversing the first database; and judging whether the piece of member code data exists in the first database.

68. The method according to claim 67, wherein updating the first database comprises: CA 3138147 Date reçue / Received date 2025-11-09 30 where the piece of member code data exists in the first database, merging member data to which the piece of member code data corresponds in the second database with member data to which the piece of member code data corresponds in the first database; where the piece of member code data does not exist in the first database, adding to the first database the piece of member code data and the member data to which the piece of member code data corresponds.

69. The method according to claim 68, wherein merging the member data to which the piece of member code data corresponds in the second database with member data to which the piece of member code data corresponds comprises adding the member data to attribute fields of the first database.

70. The method according to claim 68, wherein adding to the first database the piece of member code data and the member data to which the piece of member code data corresponds comprises: adding the member data to primary key and attribute fields of the first database; and recording and storing the member code.

71. The method according to claim 67 further comprising: updating the starting point and the cutoff point; taking the cutoff point as a new starting point, and selecting a new cutoff point after the cutoff point; selecting member behavior information between the new starting point and the new cutoff point; creating a new second database on the basis of the member behavior information between the new starting point and the new cutoff point; and executing the comparison of the member data. CA 3138147 Date reçue / Received date 2025-11-09 31 72. The method according to any one of claims 61 to 71 further comprising judging whether member codes in the second database have been stored.

73. The method according to any one of claims 61-72, wherein the member data to which the member code data corresponds comprises behavior date data and behavior fact value of the member, and wherein the behavior date data and the behavior fact value mutually correspond.

74. The method according to claim 73, wherein the member data to which the member code data corresponds comprises commodity code data and / or shop code data.

75. The method of any one of claims 73-74 further comprising extracting valid data from the member behavior information.

76. The method of claim 75, wherein the extracted valid data comprises the member code data, the behavior date data, and the behavior fact value.

77. The method according to any one of claims 73-76, wherein a corresponding logical relation is established among the member code data, the behavior date data and the behavior fact value in the first database through a JSON array.

78. The method according to any one of claims 73-77, wherein each behavior fact value indicates the frequency of a corresponding behavior as a user browses a commodity, puts an order on the commodity, or lists the commodity on a wish list.

79. The method according to claim 68, wherein merging the member data to which the piece of member code data corresponds in the second database with the member data to which the piece of member code data corresponds in the first database and updating the total database comprises: traversing each piece of behavior date data to which the piece of member code data corresponds in the second database in member data to which the piece of member code data corresponds in the first database.

80. The method according to claim 79, wherein updating the first database comprises: CA 3138147 Date reçue / Received date 2025-11-09 32 where the piece of behavior date data exists in the member data to which the member code data corresponds in the first database, replacing a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the first database, with a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the second database and updating the first database; where the piece of behavior date data does not exist in the member data to which the member code data corresponds in the first database, adding the behavior date data to which the member code data corresponds and the behavior fact value to which the behavior date data corresponds in the second database to the member data to which the member code data corresponds in the first database and updating the first database.

81. A computer program product comprising a computer readable storage medium storing computable executable instructions thereon that when executed by a computer perform the following operations: determining a starting point; selecting member behavior information from before the starting point; creating a first database on the basis of the member behavior information from before the starting point; determining a cutoff point; selecting member behavior information from between the starting point and the cutoff point; creating a second database on the basis of the member behavior information from between the starting point and the cutoff point; wherein the starting point is prior to the cutoff point, and wherein the first database and the second database each comprise member code data and member data sequenced according to the member code data; CA 3138147 Date reçue / Received date 2025-11-09 33 respectively traversing each piece of the member code data from the second database in the first database to generate a traversing result; updating the first database according to the traversing result; obtaining a third database; and classifying membership categories on the basis of the third database.

82. The computer program product of claim 81, wherein classifying the membership categories on the basis of the third database comprises assigning a label to a user according to the third database.

83. The computer program product of claim 82, wherein the label is “old category”.

84. The computer program product of claim 82, wherein the label is “new category”.

85. The computer program product of claim 82, wherein the label is “retaining category”.

86. The computer program product of claim 82, wherein the label is “repeatedly purchasing category”.

87. The computer program product according to any one of claims 81-86, wherein respectively traversing member code data from the second database in the first database comprises: executing a comparison of the member code data; selecting each piece of member code data from the second database; traversing the first database; and judging whether the piece of member code data exists in the first database.

88. The computer program product according to claim 87, wherein updating the first database comprises: CA 3138147 Date reçue / Received date 2025-11-09 34 where the piece of member code data exists in the first database, merging member data to which the piece of member code data corresponds in the second database with member data to which the piece of member code data corresponds in the first database; where the piece of member code data does not exist in the first database, adding to the first database the piece of member code data and the member data to which the piece of member code data corresponds.

89. The computer program product according to claim 88, wherein merging the member data to which the piece of member code data corresponds in the second database with member data to which the piece of member code data corresponds comprises adding the member data to attribute fields of the first database.

90. The computer program product according to claim 88, wherein adding to the first database the piece of member code data and the member data to which the piece of member code data corresponds comprises: adding the member data to primary key and attribute fields of the first database; and recording and storing the member code.

91. The computer program product according to claim 87, wherein the operations further comprise: updating the starting point and the cutoff point; taking the cutoff point as a new starting point, and selecting a new cutoff point after the cutoff point; selecting member behavior information between the new starting point and the new cutoff point; creating a new second database on the basis of the member behavior information between the new starting point and the new cutoff point; and executing the comparison of the member data. CA 3138147 Date reçue / Received date 2025-11-09 35 92. The computer program product according to any one of claims 81 to 91, wherein the operations further comprise judging whether member codes in the second database have been stored.

93. The computer program product according to any one of claims 81-92, wherein the member data to which the member code data corresponds comprises behavior date data and a behavior fact value of the member, and wherein the behavior date data and the behavior fact value mutually correspond.

94. The computer program product according to claim 93, wherein the member data to which the member code data corresponds comprises commodity code data and / or shop code data.

95. The computer program product of any one of claims 93-94, wherein the operations further comprise extracting valid data from the member behavior information.

96. The computer program product of claim 95, wherein the extracted valid data comprises the member code data, the behavior date data, and the behavior fact value.

97. The computer program product according to any one of claims 93-96, wherein a corresponding logical relation is established among the member code data, the behavior date data and the behavior fact value in the first database through a JSON array.

98. The computer program product according to any one of claims 93-97, wherein each behavior fact value indicates the frequency of a corresponding behavior as a user browses a commodity, puts an order on the commodity, or lists the commodity on a wish list.

99. The computer program product according to claim 88, wherein merging the member data to which the piece of member code data corresponds in the second database with the member data to which the piece of member code data corresponds in the first database and updating the total database comprises: traversing each piece of behavior date data to which the piece of member code data corresponds in the second database in member data to which the piece of member code data corresponds in the first database. CA 3138147 Date reçue / Received date 2025-11-09 36 100. The computer program product according to claim 99, wherein updating the first database comprises: where the piece of behavior date data exists in the member data to which the member code data corresponds in the first database, replacing a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the first database, with a behavior fact value to which the behavior date data corresponds and to which the member code data corresponds in the second database and updating the first database; where the piece of behavior date data does not exist in the member data to which the member code data corresponds in the first database, adding the behavior date data to which the member code data corresponds and the behavior fact value to which the behavior date data corresponds in the second database to the member data to which the member code data corresponds in the first database and updating the first database. CA 3138147 Date reçue / Received date 2025-11-09